1874—1960 · 2012-08-01 · at kingston, ontario, april 29, 1874. he died at bloomington, ......

n a t i o n a l a c a d e m y o f s c i e n c e s

Any opinions expressed in this memoir are those of the author(s)and do not necessarily reflect the views of the

National Academy of Sciences.

s a m u e l a l f r e d m i t c h e l l

1874—1960

A Biographical Memoir by

c . g . aBB ot

Biographical Memoir

Copyright 1962national aCademy of sCienCes

washington d.C.

SAMUEL ALFRED MITCHELL

April 29, i8j4-February 22, i960

BY C. G. ABBOT

SAMUEL ALFRED MITCHELL, son of John Cook Mitchell, was bornat Kingston, Ontario, April 29, 1874. He died at Bloomington,

Indiana, February 22, i960. His father (a tenth child) emigrated toCanada from Austell, Cornwall. His mother, Sarah Chown, born inKingston, was the tenth child of parents who emigrated to Canadafrom Devonshire. Alfred was the sixth of the ten children born tohis parents.

Well grounded in the 3 R's, at twelve years of age he passed to theKingston Collegiate Institute. With advanced standing in math-ematics, Latin, French, and German, and honored by receiving theGovernor General's scholarship in general proficiency, he went onto Queen's University. Here, through slender, he took part in col-lege sports. In his junior year came his first acquaintance with as-tronomy, which was taught by Reverend James Williamson, abrother-in-law of Sir John A. MacDonald. The subject of astronomycame very naturally after two years of mathematics with ProfessorNathan F. Dupuis and two years of physics.

Queen's then had a small wooden observatory, with two instru-ments on loan from the Royal Astronomical Society of London, atransit instrument, and a 6-inch Clark equatorial. There were twoclocks, a sidereal clock and a mean-time clock built by ProfessorDupuis, who had been an expert clock maker before he came to bea Professor of Mathematics.

SAMUEL ALFRED MITCHELL 255

At eighty, Professor Williamson, affectionately called Uncle Billy,found it pleasant and convenient to delegate the care of these instru-ments to young Mitchell. Thus he early came to know something ofthe technique of an astronomical observatory.

Mitchell received the degree of Master of Arts in 1894, and wasalso awarded the medal in mathematics. In the following session hewas appointed Tutor in Physics, continuing in charge of the observa-tory, and in the next session was appointed Instructor in Mathe-matics. But Professor Dupuis strongly recommended that Mitchellshould go to The Johns Hopkins University for graduate instructionwith the great mathematical astronomer Simon Newcomb.

Arriving at Hopkins in 1895, Mitchell found that Professor New-comb had just retired as professor, and that Thomas Craig was headof mathematics and Charles Lane Poor of astronomy. He foundCharles Lane Poor's teaching was of very high quality, and his per-sonality very attractive, so that Mitchell became inclined to adoptastronomy rather than mathematics as a life profession. He there-fore changed his plans, making astronomy his major subject, physicsfirst minor, and mathematics second minor in working for his de-gree.

Physics at Hopkins in those days took a very high place under Pro-fessor Henry A. Rowland. It used to be said, I know not how truly,that Professor Rowland was once asked in court "Who is the greatestphysicist in the world?" "I am", said he. Being asked about it later,Professor Rowland said "I could not say otherwise. I was under oath."Professor Joseph Ames, a superb teacher, was Rowland's collabora-tor; L. E. Jewell was Rowland's assistant in preparing the great chartand table of the solar spectrum. The mechanic Schneider was Row-land's able aide in perfecting the dividing engine which ruled thefamous Rowland gratings for spectroscopy.

As he was one of ten children some financial help was essential.It was most gratifying, therefore, that an assistantship in astronomywas awarded Mitchell in his second year at Hopkins. His duties

256 BIOGRAPHICAL MEMOIRS

consisted mainly in caring for the transit instrument and clocks inthe little observatory back of the physics laboratory, and the 9%-inchrefractor in the dome of the laboratory roof.

In those years, when the fine Rowland diffraction gratings werebecoming available, along with photographic dry plates of ever-in-creasing sensitiveness and spectral range, astronomic spectroscopywas flowering. Rowland's invention of the concave diffraction grat-ing opened new possibilities. His great and highly accurate map ofthe solar spectrum demanded equally accurate knowledge of thespectra of the chemical elements, in order to discover the constitu-tion of the sun. In Europe Kayser and Runge, Eder and Valenta, inEngland Sir Norman Lockyer and Alfred Fowler; in America HenryCrew and others, were gradually filling this gap by their laboratorystudies of the arc and spark spectra of the elements. A few years be-fore, George E. Hale at Chicago, and Henri Deslandres in Paris, hadindependently conceived the device of the spectroheliograph for de-lineating areas of activity of single chemical elements on the sun.Hale had persuaded the rich merchant, Charles T. Yerkes, to pro-cure from Alvan Clark and Sons what is still the largest refractingtelescope in the world, and to found the Yerkes Observatory, withHale the Director.

In these stimulating times, and being at the home of the Rowlandgratings, Mitchell took as the subject of his doctoral thesis the use ofthe concave diffraction grating for photographing the spectra of thestars. Before that, stellar spectra had been taken exclusively withprisms mounted in slit spectrographs, with the slits fed by telescopes.At the start Mitchell also used a slit at the focus of the o^-inch tele-scope. But he was soon led to dispense with the slit and throw thestar image directly into the focus of a concave grating. A special grat-ing of 6 inches aperture with lines 2 inches high was ruled on theRowland dividing engine. Elected to Phi Beta Kappa in 1897, Mitch-ell was awarded the degree of doctor of philosophy in 1898, and hisdissertation, including a theoretical discussion of the amount of astig-matism of a concave grating, used as described, was published in theAstrophysical Journal.


Though he had received a fellowship at Hopkins in 1897 and itwas continued by courtesy in 1898, Mitchell felt that it would be bet-ter to attempt to continue his research in astronomy, and he becamea research student at Yerkes Observatory, which had been dedicatedthe year before. He says in an autobiographical paper written in hislast years:

"Looking back over my life, I believe the year spent at Yerkes wasone of the finest years of my whole life, for I was able to imbibe amodicum of the enthusiasm for research found there, especially inmy good friend of many years, Professor E. E. Barnard." He goeson to say: "The staff of the Yerkes Observatory, so soon after itsopening, was very small. . . . Its limited budget was supplementedfrom time to time by gifts from Mr. Hale himself, who actually wasnot in residence through the winter months. Professor Edwin B.Frost, who later became director of the Observatory, during the win-ter was in residence at Dartmouth College. . . . The only senior as-tronomer was Professor Barnard, though Professor S. W. Burnhamcame up from Chicago on two nights of each week in order to makemeasures of double stars with the great 40-inch refractor. FerdinandEllerman, who worked closely with Mr. Hale for many years, andG. W. Ritchey who was then very busy in the observatory'sbasement grinding the two-foot reflector, were both in residence1898-99. . . . The winter temperatures were frightfully low. . . . Inthe month of February 1899, the thermometer rarely got above zeroFahrenheit.... With the long nights of winter . . . one could make aseries of long exposures on stellar spectra. One night these exposureswere continued for a period of 13 hours, while Professor Barnardwas working at the same time at the 40-inch; and the temperaturesin the domes . . . fell to 260 below zero, Fahrenheit.

"In those early days there were no warm clothes heated by elec-tricity. . . . One could put on all the heavy flannels he owned andperhaps borrow a fur coat. . . . One could keep comfortable for afew hours, depending on one's age and vitality, but gradually thecold would start at one's feet and creep up, so that before long onewould be numb to the waist. . . . It has always been a surprise that


none of the astronomers has been unfortunate to stub his toe whenso numbed and weighted down with so much heavy clothing. A fall. . . from the observing floor of the 40-inch when at its highest wouldhave been catapulting a distance of about 60 feet."

In a long voyage to observe the total eclipse of 1901 in Sumatra, andin later association at Mount Wilson Observatory, Professor Barnardhas told the present writer how he used to keep awake through thoselong hours, night after night, with black coffee, while he spent manyhours of the days between in photography and examining his beau-tiful star photographs.

Mitchell's 6-inch concave grating in its box was attached to the12-inch telescope, used as guiding instrument to bring the star ob-served to focus for the grating. With the small staff in residence,he was able to use the telescope as many nights per week as hewished. The spectrographs of stars he obtained gave interesting andpromising results. Exposures were made on many of the brighterstars, and on the Orion nebula. First and second orders of stellarspectra could be photographed on the same plate, with the directimage of the star between. The results demonstrated that many starscould be observed when gratings with longer lines became available,as they did indeed later.

In June, 1899, Mitchell accepted a position as Instructor in As-tronomy at Columbia University, and in December of the same yearhe married the daughter of Professor E. T. Dumble, then StateGeologist of Texas, who was later Consulting Geologist of the South-ern Pacific Railroad, an expert in petroleum production. The issue ofthis very happy marriage is Professor Allan C. G. Mitchell, Chair-man of the Department of Physics, Indiana University.

Later promoted to Adjunct Professor, Mitchell remained at Co-lumbia fourteen years, associated during that period with Profes-sors John K. Rees, Harold Jacoby, and Charles Lane Poor, who hadbeen his teacher at Johns Hopkins. Mitchell's teaching duties com-prised an undergraduate course in descriptive astronomy for studentsof Columbia, and later for girls from Barnard College, and also a


course in geodesy throughout the whole session for third-year stu-dents, which was continued through the first semester of the fourthyear. There was also a summer camp of six weeks for the civilengineers. All of these contacts, and life in the great city, were veryagreeable to him.

In 1900 Mitchell observed his first total eclipse of the sun as amember of the U. S. Naval Observatory expedition to Griffin,Georgia, to observe the eclipse of May 28. In all, he observed tentotal solar eclipses, traveling over 100,000 miles. On many of theseexpeditions he was the guest of U. S. Naval Observatory parties, andused special apparatus which he designed for that Observatory. Hislarge book, Eclipses of the Sun, has gone through five editions, andis the foremost authority on solar eclipses, especially in regard to the"flash spectrum."

During Mitchell's fourteen years at Columbia University he devotedsuch time as he could spare from his teaching duties to the reduc-tion of the results he obtained at the total solar eclipses of 1900,1901,and 1905 on the flash spectrum. This involved measures on the posi-tions of nearly 3000 flash-spectrum lines photographed with diffrac-tion gratings, and comparing them with the positions of the muchmore numerous solar lines of Rowland, and the arc and spark lab-oratory spectra of the chemical elements published by many observersin Europe and America.

Mitchell's papers on this research were published in the Astro-physical Journal, 38:407-95 (1912), and 39:166-79 (1914). I quote hissummary from Eclipses of the Sun.1

"As the result of investigations of the chromosphere from spectrataken at the time of an eclipse, it seems safe to make the followinggeneral conclusions:

"1 . Wavelengths in chromospheric and solar spectrum are prob-ably identical.

"2. Every strong line in the Fraunhofer spectrum is found in theflash spectrum, and every strong line in the latter (with the excep-

12nd ed. N. Y., Columbia University Press, 1924 (p. 242).


tion of hydrogen and helium lines) is matched by a line in theformer.

"3. The flash spectrum may therefore be regarded as a reversal ofthe Fraunhof er spectrum.

"4. The 'flash' is not an instaneous appearance. At the beginningof totality the chromospheric lines of greatest elevation appear first,and at the end of totality remain the last.

"5. The 'reversing layer' which contains the majority of the low-level lines of the chromosphere is about 600 kilometers in height.

"6. The 'reversing layer' has no existence separate from thechromosphere.

"7. It is the densest part of the chromosphere, lying closest to thephotosphere, and it is the cause of the greatest part of the absorptionproducing the Fraunhofer lines.

"8. The 'Evershed effect' measured in sun-spots and photographsof flocculi, which exhibit vastly different aspects when photographedat various elevations above the phostosphere, proves that the shad-ings of such strong lines as H and K are caused by absorption atdifferent levels and pressures above the photosphere.

"9. The chromospheric spectrum differs greatly from the ordinarysolar spectrum in the intensity of the lines.

"10. The Fraunhofer spectrum is essentially an arc spectrum. Thechromospheric spectrum more closely resembles the spark spectrum,and its spectrum corresponds to an 'earlier type' [of star] than thesun.

"11. Especially prominent in the chromosphere are the enhancedlines relatively stronger in spark than in arc spectra of the elements.

"12. The enhanced lines ascend to greater elevations above thephotosphere than do the ordinary lines.

"13. The increased elevations cause greatly diminished pressures."14. As Saha has shown, the reduced pressures permit the ioniza-

tion of the atom. As a result the lines of ionized atoms are speciallyprominent in the flash spectrum. The enhanced lines are producedby the ionized atoms.


"15. The depth of the chromosphere is not constant."

Mitchell was fortunate in enlisting deep interest in eclipses in hisfriend Edward D. Adams of New York, founder of the ErnestKempton Adams Fellowship, which is awarded each year by Co-lumbia University for researches in the domain of pure science. Atthe total solar eclipse of June 8,1918, observed at Baker, Oregon, Mr.Adams joined the Naval Observatory expedition, and took it uponhimself to find the right man to draw and paint the corona. He choseHoward Russell Butler, a portrait painter of note, whose beautifulpainting, reproduced in Mitchell's Eclipses of the Sun, is owned bythe Hayden Planetarium of New York.

In the summers of the years 1909, 1910, and 1911 and for fifteenmonths of sabbatical leave in 1912-13, Mitchell had the opportunityof working at Yerkes Observatory. In 1905 Dr. Frank Schlesinger,later Director of Allegheny Observatory, using the 40-inch refractorat Yerkes, had demonstrated that stellar distances could be deter-mined photographically with a telescope of great focal length farmore accurately than had previously been done with visual observa-tions. This work was taken up again with the 40-inch refractor atYerkes in 1912-13 by Mitchell and Frederick Slocum. Observation ofstellar parallax was to be the major field of Mitchell's later life asDirector of Leander McCormick Observatory at Charlottesville, Vir-ginia.

Leander McCormick, of the McCormick reaper family, was bornin Rockbridge County, Virginia. In 1870 he was introduced to Sec-retary Henry of the Smithsonian Institution by a letter from GeneralR. E. Lee, who stated that McCormick "wishes to erect and providewith superior instruments an Astronomical and Physical Observa-tory in the State of Virginia." Eventually a Clark 26-inch refractingtelescope was procured by him to be installed at Washington andLee University at Lexington. But difficulties arose, and ProfessorSimon Newcomb succeeded in obtaining that telescope for the U.S.Naval Observatory at Washington.


Later, however, Mr. McCormick persevered and ordered a 26%-inch refracting telescope from Clark's. It was made by the youngerAlvan Clark, who regarded it as his masterpiece. Mr. McCormickgave $50,000 for the telescope and $18,000 to construct the dome onthe top of Mount Jefferson, near the University of Virginia at Char-lottesville. Through the efforts of Professor Charles S. Venable an en-dowment of $75,000 was raised. The Leander McCormick Observa-tory was dedicated in 1883, and at that time possessed the largest andfinest refracting telescope in the world. The first Director was Pro-fessor Ormond Stone, who came in 1882 at the recommendation ofProfessor Simon Newcomb. The budget in President Alderman'stime was $4450 a year, permitting the Director's salary to be $3000a year, leaving $1450 a year for all other purposes. But in honor ofCommodore Vanderbilt, who had contributed $25,000 toward theendowment, three scholarships of $35.00 per month for tenmonths per year were established. These scholarships were prizedand were held by many men who later won considerable distinctionin astronomy.

Professor Ormond Stone remained Director for thirty years. In1913 Mitchell accepted the post as his successor. He was retired asDirector Emeritus in 1945. Mitchell keenly realized how he was ham-pered by the paucity of funds for research, and since he was unable toobtain a substantial increase from the University of Virginia, through-out his tenure as Director he devoted much time and effort to ob-taining support from outside sources. With his pleasing personalityand obvious competence and devotion for his researches, he wasquite successful, though he never received such large endowments, inthe hundreds of thousands or millions, as have sometimes beenmade.

Considering what might be his best opportunity for research,though at a disadvantage compared to the long-focus photographictelescopes, by exposures employing color screens he felt that with theexcellent 261

/4-inch refractor he might yet compete with them in ac-curacy of observation of stellar parallaxes to determine distances of


the stars. In the year 1913, when Mitchell went to Virginia, DirectorW. W. Campbell of Lick Observatory had written: "A quarter of acentury ago we knew the distances on not more than two-score stars. . . even today we possess reliable knowledge of the approximate dis-tances of not over a hundred stars." There was a crying need forknowing the distances of not hundreds but thousands of stars.

With a small balance available from the meager annual funds ofthe Observatory, Mitchell ordered a double-slide photographic platecarrier, and adjusted it to the refractor. Fortunately Columbia Uni-versity awarded him the Ernest Kempton Adams research fellow-ship in July, 1914. With this $1250 he was able to add Charles P.Olivier and Harold L. Alden to the McCormick staff, and the pho-tographic parallax work was started with enthusiasm.

For certain reasons it is not desirable to photograph for parallaxwithin two hours of midnight. So there were four hours availablefor some other worth-while researches. The Observatory had an ex-cellent micrometer which was used by Olivier for double-star meas-ures. Mitchell and Alden turned to a program on long-period variablestars which eventually grew to large proportions.

Neglecting, for simplicity's sake, the effect of the motion of thewhole solar system among the stars, we see that the earth by its mo-tion about the sun occupies positions approximately 186 million milesapart at six-month intervals. Almost all of the stars are so far awaythat this enormous distance makes no appreciable angular displace-ment in their apparent positions. But some thousands of stars are nearenough to be displaced between 0.05 and 0.2 seconds of arc, andthese displacements can be measured. Hence if a series of photographsis made at intervals of six months, such "nearby" stars will be foundby measurements of great accuracy to be displaced with respect to thestellar background. Knowing in this way their angular displace-ments, their distances may be computed by essentially the samemethod by which a surveyor might fix the distance of a tree beyonda river.

Such was the nature of the main program that Mitchell and his


aides carried on at Leander McCormick Observatory after 1914. In1915 he was assured of the Adams fellowship for another year, buthad no assurance of sufficient funds beyond 1916. He had found noprospect of help from the University, and had even been discouragedthere from any hope from the McCormick estate. But within a monthof his interview at the University a telegram came summoning himto Boston. When he arrived, he arranged with Robert H. McCor-mick, Jr., in the absence of his father for the day, to go out toHarvard College Observatory to call on Professor E. C. Pickering.That good and able man escorted them about the observatory him-self, and inquired what Mitchell was working on. Professor Picker-ing read a letter he had received from the astronomer Innes in SouthAfrica, asking if it might be possible to buy or borrow the 2654-inchtelescope at McCormick Observatory. Professor Pickering then wenton to tell how, as a young man just come to Harvard and in need offunds, some friends had guaranteed him $5000 yearly for five years.Said he: "Mr. McCormick, Mr. Mitchell is in urgent need to carryon his researches. I would recommend that your family treat himin a manner similar to the way my friends treated me a number ofyears ago."

In an interview with Robert Hall McCormick the next day, thatsuggestion bore fruit. A check for $2000 reached Mitchell in July,1915, and thereafter a close friendship followed between Mitchell andthe McCormicks, father and son. It was continued through LeanderMcCormick-Goodhart until Mitchell's death.

About this time Dr. Walter S. Adams of Mount Wilson Observa-tory discovered the spectroscopic method for determining stellarparallaxes. He found, by comparing known trigonometric parallaxeswith the spectra of certain types of stars, that certain sensitive spec-tral lines showed alterations which could be correlated with the ab-solute brightness of the star observed. Then, knowing by stellar pho-tometry the apparent brightness of the star, the distance could becomputed at once, since the apparent brightness falls off as the squareof the distance.


When the Adams fellowship from Columbia was soon to expire,Mitchell asked for a further renewal. His friend Edward DeanAdams told him that it was against policy to renew more than once,but that, as Mitchell made so strong a plea, he would consult withProfessor George E. Hale during his approaching visit to Pasadena.Hale advocated granting Mitchell's request, for, though the spectro-scopic method of determining parallaxes was available for certaintypes of stars, no matter how distant, that are bright enough to yieldsatisfactory photographic spectra, this new method needed to be but-tressed by as many trigonometric parallaxes as possible. These wereneeded for accuracy, and if possible to widen application of this spec-troscopic method to include more spectral types of stars. As a resultMr. Adams decided to establish a special fellowship of $1000 yearlyfor five years.

The exacting measures of star positions on the McCormick pho-tographic parallax plates was taken up as an Ernest Kempton Adamsresearch. Two Repsold measuring machines were loaned by Colum-bia University. With grants from the Henry Draper Fund of theNational Academy an especially suitable large measuring machineby Gaertner of Chicago was procured. According to the bibliogra-phy appended to this memoir, 2001 stellar parallaxes from McCor-mick Observatory were published prior to the year 1950. Compari-sons with parallaxes duplicated by other observatories indicated thatin accuracy McCormick parallaxes were on a par with the best.

Professor E. C. Pickering of Harvard had long been interested inpromoting measurements of stellar photometry, the relative bright-ness of stars. With a small grant from the Rumford Fund of theAmerican Academy of Arts and Sciences he had procured five iden-tical photometers and had secured the use of them on the large re-fracting telescopes of Harvard, Princeton, McCormick, Lick, andYerkes Observatories. The object was to measure visually the mag-nitudes of faint stars surrounding certain long-period variable starsof magnitudes ranging from 12 to 16, so that these newly measuredfaint stars might serve as a scale for all variable star observers. Part


of the work planned had been done, but about 1905 progress uponit stopped for various reasons.

In the year 1917 Mitchell visited Professor Pickering at Harvard,and among other subjects they discussed possible completion of thisproject. It appeared to be a useful occupation for the 261

/4-inch re-fractor between the hours devoted to parallax photography. Mitchelland Alden together measured the 12th, 14th, and 15th magnitudestandard stars, and repeated the measures on the 16th magnitudeswhich had already been observed at Yerkes and Lick.

But new and highly accurate methods of photographic photome-try came into use, and the careful visual work of many men andmany years on the Pickering plan, though published, never cameto be extensively appreciated. Yet the experience gained by Mitchelland Alden in stellar photometry became the basis for new McCor-mick projects of great value. They became especially interested in thevariable stars of long periods of from 100 to 400 days. There wereavailable at the time a series of charts designed and published byFather Hagen, astronomer at Georgetown Observatory, showing avariable at the center of the chart, and around it neighboring starsrepresented by circles of differing diameters to distinguish theirmagnitudes. Some observations made at McCormick were com-pared with the Hagen charts, and stars fainter than those printed onthe charts were added in pencil, but their magnitudes were unknown.

It seemed best to improve the basis for their long-period variableobserving. A collection of 200 photographs was obtained from Har-vard and Yerkes covering the regions of as many long-period vari-ables as possible. These regions were then photographed with the261

/4-inch refractor, and each photograph marked in ink with theHarvard magnitude of each comparison star known. When Mitchelland Alden began to use these photographs they found discrepanciesin magnitudes, and for stars fainter than 12th magnitude decided dis-cordances. So they understook to reobserve visually these magnitudeswith the wedge photometer and the 26 54-inch refractor. Then theyformed a scale of steps of tenths of magnitudes for all companion


stars of a sequence, carrying the scale to about one-half magnitudebelow the faintest phase of the variable star. Details of the procedurewere explained in Volume VI of the Leander McCormick Publi-cations.

As the limit of the 26% refraction for visual observation is 16thmagnitude, and many of the long-period variables at faintest reachthat magnitude, Mitchell was invited by Director W. S. Adams ofMount Wilson Observatory to use the 60-inch reflector on certainnights to perfect the scale.

In 1934 McCormick Observatory was presented with a 10-inchCooke photographic telescope, as a gift from Mount Wilson Observa-tory, the Carnegie Institution of Washington, and the Carnegie Cor-poration. Using ordinary plates, it portrays the stars in so-calledphotographic light. With yellow-sensitive plates and a yellow colorfilter, the portrayal is "photo-visual." When a thin prism is placedover the objectives it becomes a "prismatic camera" yielding as manyas 2000 spectra on an 8 x 10 plate. This instrument was highly use-ful in the long-period variable program of Mitchell and Alden.

The most accurately determined magnitudes in the whole sky arethose about the North Pole called the North Polar Sequence. Withthe 10-inch Cooke telescope photographs including the North PolarSequence and the region of a long-period variable of equal altitudeabove the horizon could be made on the same plate. Measures witha microphotometer on 10-inch and on 26% -inch plates furnishedaccurate magnitudes so acquired. Yellow-sensitive plates exposedthrough yellow-color filters made the magnitudes "photo-visual."Visual magnitudes plotted against these photo-visual ones fell onstraight lines, and it was found that the two sorts are indeed indis-tinguishable the one from the other in McCormick observations.

After the work was well along, the McCormick magnitudes wereput on a photograph of the long-period variable region and sent tothe chairman of the Chart Committee of the American Association ofVariable Star Observers. He draughted a tracing from which blue-prints were made for the members of the Association. In the course


of twenty years of this friendly cooperation over 8000 charts havebeen derived, and are adapted for the 450 stellar regions that areunder continued observation by members of the Association. At theinternational conferences of variable-star observers at Harvard in1932, Paris in 1935, and Stockholm in 1938, Director Mitchell wasinvited to preside.

This and other researches were sandwiched between the hours ofstellar parallax work under Mitchell's directorship at Leander Mc-Cormick Observatory. The parallax work, his leading research, com-prised observation of: (1) The brightest stars of all spectral typesdown to magnitude 5.0 and some even fainter. (2) Stars of notableproper motions of 0.5 seconds of arc per year or less. (3) Double stars.As already stated, parallaxes of 2001 stars were published by Mitchellby the year 1950.

Eleven observatories with large telescopes located in both hemi-spheres observed trigonometric stellar parallaxes. There are manyduplicated stars. Comparisons indicate that McCormick parallaxesare on a par with the best. Great numbers of spectroscopic parallaxeshave also been published. To give them a proper basis for convertingabsolute luminosity determinations into stellar distances they requirenumerous trigonometric parallaxes for each spectral type employed.As sensitive lines were later discovered in spectral types not at firstfound suitable, the call went out for new trigonometric parallaxesnot formerly on parallax programs. This led to close cooperation be-tween Directors Adams of Mount Wilson and Mitchell of LeanderMcCormick.

Among the stars of special spectrum types whose trigonometricparallaxes were desired were the so-called giants, like Betelgeuse,which are mostly at very great distances. Also desired were parallaxesfor the dwarf M-type stars, which may be near, and if so will bestars of large proper motion. The prismatic 10-inch camera at Mc-Cormick in the skillful hands of Dr. A. N. Vyssotsky can providespectroscopic parallaxes and can detect the difference between giantM and dwarf M spectral types. With 2000 spectra of stars on one


plate it takes time to ferret out the stars of special interest. Whendetected, a dwarf M must be sought on the records of other observ-ers, and its proper motion determined before deciding to put it onthe McCormick parallax list for observation. On four summer visitsto Mount Wilson, Director Mitchell was able to use the 6o-inch re-flector to secure spectra of faint stars on the McCormick program.

This spectroscopic adjunct to the parallax research called forproper motions, and led to the publication, in Volume 10 of Mc-Cormick Publications, of the proper motions of over 70,000 stars ascatalogued by Dr. and Mrs. Vyssotsky. The catalogue covers seven-eighths of the whole sky from pole to pole, and yields the followingimportant astronomical results: (1) The motion of the sun has beendetermined with reference to stars of different colors and differentmagnitudes. (2) The period of rotation of the Galaxy (at about200 million years) is given with higher precision. (3) The constantsof the precession of the equinoxes are fixed with great exactitude.

Mitchell's scholastic and honorary degrees included: M.A., Queen'sUniversity, Ontario, 1894; Ph-D., The Johns Hopkins University,1898; L.L.D., Queen's University, 1924; L.L.D., University of West-ern Ontario, 1940.

He was a member of the following societies: National Academy ofSciences (elected in 1933); American Association for the Advance-ment of Science (Vice-President, 1921); American AstronomicalSociety (Vice-President 1925-27); American Philosophical Society(Fellow); American Association of University Professors (President1934, 1935); American Academy of Arts and Sciences (Fellow).

He received the James Craig Watson Medal of the National Acad-emy of Sciences in 1948.


KEY TO ABBREVIATIONS

Astron. J. = The Astronomical JournalAstron. N. = Astronomische NachrichtenAstrophys. J. = Astrophysical JournalColumbia Contr. Phil. Psych. = Columbia University, Contributions to Philoso-

phy and PsychologyHand. Astrophys. = Handbuch der AstrophysikJ. Roy. Astron. Soc. Canada = The Journal of the Royal Astronomical Society of

CanadaLeander McCormick Pub. = Leander McCormick Observatory PublicationsMem. Am. Acad. Arts Sci. = Memoirs of the American Academy of Arts and

SciencesMonthly Notices Roy. Astron. Soc. = Monthly Notices of the Royal Astronomi-

cal SocietyObs.^The ObservatoryPop. Astron.=Popular AstronomyPop. Sci. Monthly = Popular Science MonthlyProc. Am. Phil. Soc. = Proceedings of the American Philosophical SocietyPub. Astron. Soc. Pac. = Publications of the Astronomical Society of the PacificPub. Nat. Geog. Soc. Eclipse Pub. = National Geographic Society, Solar Eclipse

SeriesSci. Am. = Scientific AmericanSmith. Inst. Pub. = Smithsonian Institution PublicationsYerkes Pub. = Yerkes Observatory, Publications

B I B L I O G R A P H Y

1898

Notes on the Concave Grating. Astrophys. J., 8:102.With. Charles Lane Poor. The Concave Grating for Stellar Photography.

Monthly Notices Roy. Astron. Soc, 58:2o.i-95.

1901

Examinations of the Pleiades and Eros Plates Taken with the Crossley Re-flector of the Lick Observatory. Astrophys. J., 13:48.

The Government Eclipse Expedition. Science, 84:226.Total Eclipse of the Sun. Science, 14:802.Focal Singularities of Plane Gratings. Astrophys. J., 14:331.The Flash Spectrum, Sumatra Eclipse, May 18, 1901. Science, 15:257.

SAMUEL ALFRED MITCHELL 27I

1902

The Flash Spectrum, May i8, 1901. Wave-length Determinations andGeneral Conclusions Regarding the "Reversing Layer." Astrophys. J., 15:97-

1903

The New Gases, Neon, Argon, Krypton, and Xenon in the Chromosphere.Astrophys. J., 17:224.

The Sumatra Eclipse, 1901. Spectroscopic Study of the Flash Spectrum.New Gases in the Sun. Columbia Contr. Phil. Psych., 20, 42 pp.

1904

Comet 1903 Borrelly and Light-pressure. Astrophys. J., 20:63.

1905

Purposes and Plans of the Solar Eclipse Expedition of August, 1905. Sci-ence, 21:918.

Preliminary Account of Flash Spectrum taken 1905, August. MonthlyNotices Roy. Astron. Soc, 66:326-28.

1906

Preliminary Results of United States Naval Observatory Eclipse Expe-dition in 1905. Astrophys. J., 23:128.

1907

"Knots" in the Rings of Saturn. Sci. Am., 97:376.

1908

An Eclipse Expedition to Spain. Pop. Sci. Monthly, 68:551-63.

1909

Photographing a Star Spectrum. Sci. Am., 101:495.Seven Spectroscopic Binaries. Astrophys. J., 30:239-42.Peculiar Behavior of Morehouse's Comet. Sci. Am., 100:26.

1910

A Great Open-air Telescope. Pop. Astron., 18:296.Other Worlds in Space. Sci. Am., 102:160.Telescope Lenses and How to Test Them. Sci. Am., 102:363.An Interesting Spectroscopic Binary. Pop. Astron., 18:581.


19II

The Radial Velocities of 96 Hercules. Science, 34:529.

1912

Radium and the Chromosphere. Astron. N., 192:265-70.Wave-lengths of the Chromosphere from the Spectra Obtained at the

1905 Eclipse. Astrophys. J., 38:407-95.Brooks' Comet. Sci. Am., 105 =299.

Is Radium in the Sun? Pop. Astron., 211321-31.

1914

The Depth of the Reversing Layer. Astrophys. J., 39: 166-79.With F. Slocum. Stellar Parallaxes from Photographs Made with the 40-

inch Refractor of the Yerkes Observatory. Astron. N., 197:81.

1915

Stellar Parallax Work at the McCormick Observatory. Astrophys. J., 42:263-70.

1916

Systematic Observations of Meteors. Sci. Am., 113:48.Preliminary Parallax of Barnard's Star with the 26-inch Refractor of the

Leander McCormick Observatory. Astron. J., 30:75.

1917

With F. Slocum. Parallaxes of Forty-two Stars from Photographs Madewith the Forty-inch Refractor. Yerkes Pub., 4(pt. i):5-27-

1919

The Total Solar Eclipse of 1918. Natural History, 19:244-71; Leander Mc-Cormick Pub., 2(pt. 6) :2

1920

Parallaxes of 260 Stars Derived from Photographs. Leander McCormickPub., 3:1-677.

The Parallax of Krueger 60. Astron. J., 33:177-79.


1921

With H. L. Alden. Observations of SS Aurigae and Observations of FaintComparison Stars. Astron. J., 33:87-91.

1922

Trigonometric Parallaxes of Stars of A and B Types; 19 B Stars and 50A Stars. Pub. Astron. Soc. Pac, 341254-57.

Trigonometric Parallaxes of Twenty-two Cepheids Measured at the Mc-Cormick Observatory. Obs.,

1923

Eclipses of the Sun. N. Y., Columbia University Press, xviii -f- 425 pp.How the Sun's Atmosphere Is Studied at Eclipses and the Interpretation

of the Results through the Aid of Modern Physics. Pop. Astron., 31:652.

1924

With H. L. Alden. Photometric Standards of Faint Standard Stars Meas-ured Visually at Harvard, Yerkes, Lick and McCormick Observatories.Mem. Am. Acad. Arts Sci., 4:209-307.

Eclipses of the Sun, 2nd ed. N . Y., Columbia University Press, xvii + 452pp.

1925

Trigonometric Parallaxes of Two Hundred Stars of Large Proper Motion.Pub. Astron. Soc. Pac, 36:185-91.

1926

McCormick Parallaxes of Stars of B Type. Pub. Astron. Soc. Pac, 38:101-6.

Trigonometric Parallaxes of Forty-seven Stars. Astron. J., 36:143-44.With H. L. Alden. The Photometric Scale of the Leander McCormick

Observatory. Monthly Notices Roy. Astron. Soc, 86:356-60.

1927

With C. G. Abbot. The Fundamentals of Astronomy. N . Y., Van Nos-trand, ix + 307 pp.

1928

With C. P. Olivier and H. L. Alden. Trigonometric Parallaxes of Four


Hundred and Forty Stars. Leander McCormick Pub., 4:349 pp.The Distances of the Stars. Proc. Am. Phil. Soc, 67:267-86.

1929

Eclipses of the Sun. Hand. Astrophys., 4:231-355.The Relation between Corona and Sun-spots. Pop. Astron., 37:192-98.Systematic Errors of Parallaxes. Pop. Astron., 37:271.Trigonometric Parallaxes of Thirty-one Stars. Astron. J., 39:83-84.Atlas Stellarum Variabilium, Series VII. Monthly Notices Roy. Astron.

Soc, 89:654-58.Trigonometric Parallaxes of Thirty-one Stars. Astron. J., 39:89-90.

1930

Recent Achievements in Measuring Stellar Distances. Scientia, 48:217-26,291-301.

The Spectrum of the Chromosphere. Astrophys. J., 71:1-61.Heights in the Chromosphere. Astrophys. J., 72:146-86.

1932

Eclipses of the Sun, 3rd ed. N. Y., Columbia University Press, xvii + 490pp.

Spectroscopic Discoveries at the Recent Total Eclipse. Proc. Am. PhiLSoc, 71 =343-48.

1933With E. T. R. Williams. Relative Distribution and Abundances in the

Lower Chromosphere. Astrophys. J., 77:1-43.Heights in the Chromosphere from Fixed and Moving Plates. Astrophys.

J., 77:i57-85-Trigonometric Parallaxes of Twenty-three Stars. Astron. J., 43:63.

1934Systematic Errors of Parallaxes. Astrophys. J., 80:200-228.Eclipses of the Sun, 4th ed. N. Y., Columbia University Press, xvii -f- 520

pp.

1935The Magnitudes of 6284 Stars in the 350 Regions of Long-period Vari-

ables. Science, 82:624.


With C. A. Wirtanen. A Comparison Sequence for Nova Herculis. As-t ron j . , 45:31-32.

Observations of Nova Persei (1901). Leander McCormick Pub., 6:51-52.Observations of Nova Cygni (1920). Leander McCormick Pub., 6:161-62.Observations of Nova Tauri (1927). Leander McCormick Pub., 6:65.With C. A. Wirtanen. Observations of Nova Herculis. Astron. } . , 40:31.With H. L. Alden et al. Observations of Long Period Variables. Leander

McCormick Pub., 6(pt. 1) 3-197.Magnitudes and Coordinates of Comparison Stars Made with the 26-inch

Refractor of the Leander McCormick Observatory. Leander McCor-mick Pub., 6(pt. 2) :i

I936

Eclipses of the Sun. Hand. Astrophys. 7:382-408.The Total Solar Eclipse Observed on Canton Island. Pub. Astron. Soc.

Pac, 50:23-30.

1938

Discoveries from Solar Eclipse Observations. Smith. Inst. Pub., 3453:145-67.

With an Astronomer on an Eclipse Expedition. Proc. Am. Phil. Soc., 79:341-60.

With C. A. Wirtanen. A Comparison Sequence for Nova Lacertae.Monthly Notices Roy. Astron. Soc, 99:40-41.

1939Nature's Most Dramatic Spectacle. Pub. Nat. Geog. Soc. Eclipse Pub.,

361-94.With C. A. Wirtanen. Magnitudes and Comparison Stars of Fifty Long-

period Variables. Leander McCormick Pub., 9:(pt. 5) :59~88.

1940

With D. Reuyl. The Trigonometric Parallaxes of 650 Stars. Leander Mc-Cormick Pub., 8, 749 pp.

1941

With A. N. Vyssotsky and P. van de Kamp. The Trigonometric Parallaxesof Thirty-seven Stars. Astron. J., 49:129-31.


1942

Early American Astronomers. J. Roy. Astron. Soc. Canada, 36:345-60.

1944With C. Beddow. Trigonometric Parallaxes of One Hundred and Thirty-

one Stars. Astron. J., 51:93~96.

1945With D. Watson. Trigonometric Parallaxes of 88 Stars. Astron. J., 51:

182-84.

1946

Chromospheric Spectrum from Ten Eclipse Expeditions. Astrophys. J.,105:1-35; Leander McCormick Pub., 9(pt. 18) :2i3~47, 1947.

1947

The Observation of Ten Total Eclipses. Leander McCormick Pub., 9:(pt.17) =203-11.

1948

With A. H. Joy. Spectroscopic Observations of 90 Stars. Astrophys. J., 108:234-36.

1949

With C. Beddow and D. Watson. Trigonometric Parallaxes of 28 Stars,Determined by Photography with the 26-inch McCormick Refractor.Astron. J., 54:95.

1951

Eclipses of the Sun, 5th ed. N . Y., Columbia University Press, xv + 445pp.

1958

Widi D. Reuyl, C. M. Anderson, et al. Trigonometric Parallaxes of 432Stars. Leander McCormick Pub., I4:(pt. 1).

1874—1960 · 2012-08-01 · at kingston, ontario, april 29, 1874. he died at bloomington, ......

Documents